Emodin inhibits migration and invasion of DLD-1 (PRL-3) cells via inhibition of

Emodin inhibits migration and invasion of DLD-1(PRL-3)cells via inhibition of PRL-3phosphatase activity

Young-Min Han a ,c ,Su-Kyung Lee a ,Dae Gwin Jeong b ,Seong Eon Ryu b ,Dong Cho Han a ,c ,Dae Keun Kim d ,?,Byoung-Mog Kwon a ,c ,?

a

Laboratory of Chemical Genomics and Biology,Korea Research Institute of Bioscience and Biotechnology,1125Gwahakro,Yoosunggu,Daejeon 305-600,Republic of Korea b

Proteomics Research Center,Korea Research Institute of Bioscience and Biotechnology,1125Gwahakro,Yoosunggu,Daejeon 305-600,Republic of Korea c

University of Science and Technology,Daejeon 305-333,Republic of Korea d

College of Pharmacy,Woosuk University,Samrye 565-701,Republic of Korea

a r t i c l e i n f o Article history:

Received 10September 2011Revised 29October 2011Accepted 2November 2011

Available online 9November 2011Keywords:Emodin Metastasis PRL-3Migration Invasion

a b s t r a c t

Anthraquinones have been reported as phosphatase inhibitors.Therefore,anthraquinone derivatives were screened to identify a potent phosphatase inhibitor against the phosphatase of regenerating liver-3(PRL-3).Emodin strongly inhibited phosphatase activity of PRL-3with IC 50values of 3.5l M and blocked PRL-3-induced tumor cell migration and invasion in a dose-dependent manner.Emodin res-cued the phosphorylation of ezrin,which is a known PRL-3substrate.The results of this study reveal that emodin is a PRL-3inhibitor and a good lead molecule for obtaining a selective PRL-3inhibitor.

ó2011Elsevier Ltd.All rights reserved.

Tumor progression is a complex process that includes malig-nant transformation,proliferation,invasion,and metastasis of can-cer cells.Cancer cell invasion and metastasis are critical processes and are the major causes of death for cancer patients.1,2Approxi-mately 90%of all cancer deaths arise from the metastatic spread of tumors.3To overcome this problem,a greater understanding of the biology of metastasis and the development of novel therapeu-tics that speci?cally target metastasis and metastatic progression are required.4–6Colorectal cancer (CRC)is the third most com-monly diagnosed cancer in males and the second in females,and caused over 1.2million new cancer cases and approximately 608,700deaths in 2008.7CRC is also the third most common type of cancer and cause of cancer-related deaths in the United States in men and women with an estimated 102,900new cases in 2010.8Many CRC patients suffer from the unexpected development of metastasis,after the curative resection of their primary tumors,and approximately 50%develop liver metastasis at some point dur-ing their disease.9

The phosphatase of regenerating liver family (PRL-1,-2,and -3)constitute three closely related intracellular enzymes that possess the protein-tyrosine phosphatase (PTP)active site.10Saha et al.

0960-894X/$-see front matter ó2011

Elsevier Ltd.All rights reserved.doi:10.1016/j.bmcl.2011.11.008

?Corresponding authors.Tel.:+82428604557;fax:+82428612675(B.-M.K.).

E-mail addresses:dkkim@mail.woosuk.ac.kr (D.K.Kim),kwonbm@kribb.re.kr (B.-M.Kwon).

have reported that PRL-3is expressed at high levels in nearly all metastatic lesions that are derived from CRC,regardless metastasis site.However,little or no PRL-3expression is observed in normal colon and non-metastatic primary cancers.11,12PRL-3promotes

DLD-1 (PRL-3)

15hr

20hr

Control

120

3040

100

120

D

324Y.-M.Han et al./Bioorg.Med.Chem.Lett.22(2012)323–326

cancer invasion,migration,growth,and angiogenesis via dephos-phorylation that is catalyzed by the catalytic domain or localiza-tion to plasma membrane that is directed by the C-terminal prenylation motif.13–15A previous study has reported that phos-phatase activity of PRL-3plays an essential role in tumor metasta-sis.16These studies suggest that PRL-3may be an ideal therapeutic target for drug development to inhibit the spread of colorectal cancer.

In the search for potential PRL-3inhibitors from natural sources, we previously reported that anthraquinone and bi?avonoids inhib-ited the activity of PRL-3.17,18In the current study,we screened com-mercially available anthraquinones and bi?avonoids using a DiFMUP(6,8-di?uoro-4-methylumbelliferyl phosphate)assay. Emodin was identi?ed as a potent PRL-3inhibitor.Anthraquinones represent a large family of compounds with diverse biological properties.Emodin(1,3,8-trihydroxy-6-methylanthraquinone)is an anthraquinone derivative and inhibits cellular proliferation,induces apoptosis,and prevents metastasis.19Its capabilities are mediated by the inhibition of tyrosine kinases,phosphoinositol3-kinase(PI3K),protein kinase C(PKC),NF-kappa B(NF-kappaB), and mitogen-activated protein kinase(MAPK)signaling cascades.19 This study is the?rst report to show that emodin blocks tumor cell migration and invasion via PRL-3inhibition.

To identity the PRL-3inhibitor,anthraquinones and bi?avo-noids were tested against PRL-3protein using a DiFMUP assay.17,20 As previously reported,anthraquinones and bi?avonoids inhibited PRL-3phosphatase activity,17,18with IC50values of that were high-er than20l M.However,emodin strongly inhibited PRL-3phos-phatase activity in a dose-dependent manner with an IC50value of3.5l M(Fig.1A and B).We selected emodin as a good lead mol-ecule for PRL-3inhibitor.Even though we did not test the inhibi-tory activity of emodin against PRL-1and2,emodin may not selective inhibitor against PRL-3,because PRL-1,PRL-2,and PRL-3 share76–87%amino acid sequence identity.21

To examine the ability of emodin as a PRL-3inhibitor in tumor cells,we prepared PRL-3-expressed DLD-1colon tumor cells[DLD-1(PRL-3)]because PRL-3induces the invasion of human tumor cells.22,23To investigate the role of over expressed PRL-3in cancer cell metastatic processes,we used cell migration and invasion as-says using Transwell migration and Matrigel invasion chambers, respectively.PRL-3-expressed cells migrated faster than the con-trol cells in the cell migration assay(Fig.2A).After20h,the cell migration was saturated.The cell invasion was measured at20h when the migration control wells were saturated.We observed sig-ni?cant invasion with the DLD-1(PRL-3)colon tumor cells com-pared with the parental cells(data not shown).We investigated whether the PRL-3phosphatase inhibitor(emodin)modulated PRL-3-dependent migration and invasion.DLD-1(PRL-3)colon tu-mor cells were incubated in different concentrations of emodin for

Control

120 3040

s 100

120

140

DLD-1(PRL-3)

0 10 20 30 40 (μM)

rescue of ezrin after emodin treatment in DLD-1or

cells that were treated with emodin were prepared

Tris,pH7.0,150mM NaCl,5mM EDTA,1%Triton X-100, Y.-M.Han et al./Bioorg.Med.Chem.Lett.22(2012)323–326325

15h at37°C in a humidi?ed atmosphere of5%CO2.Cell migration assay was performed as described previously using8.0l m pore size Transwell inserts(Costar Corp.,Cambridge,MA)with some modi?cations.24As shown in Figure2B and C,emodin inhibited the migration of DLD-1(PRL-3)cells in a dose-dependent manner with an IC50value of28l M.After the cells were treated with 40l M of emodin,more than80%of the cell migration was blocked. To clarify whether the inhibitory effect of emodin on the migration of DLD-1(PRL-3)cells was due to a cytotoxic effect,the compound was tested for its effects on the cell proliferation.Emodin mildly inhibited proliferation of DLD-1(PRL-3)cells up to50l M.These results indicate that emodin signi?cantly inhibits cell migration and invasion without cytotoxicity(Fig.2D).

The invasive behavior of cancer cells is accompanied by an in-creased cell movement,suggesting that cell migration blockers can inhibit tumor invasion.The effect of emodin on DLD-1(PRL-3)cell invasion was examined using a Matrigel-coated8.0l m?lter invasion chamber assay(BD Biosciences CA).24The cells were trea-ted with emodin at different concentrations(10–40l M)for20h, and10%FBS was used as a chemoattractant.Emodin inhibited the invasion of DLD-1(PRL-3)cells in a dose-dependent manner with an IC50value of35l M(Fig.3A and B).In particular,cell invasion was strongly inhibited by40l M emodin(approximately$70%).

Ezrin,which is a linker protein between the plasma membrane and cytoskeleton,has been reported to be a direct target of PRL-3. For the rescue of phosphorylation on ezrin,we used a-ERM pThr 567antibody because phosphor-ERM antibody is generally used to detect phosphor-Ezrin Thr567.25In order to con?rm the inhibi-tion of PRL-3by emodin,we examined its ability to rescue the phosphorylation of a-ERM-Thr567.When emodin treated DLD-1 (PRL-3)cells was incubated for15h,the phosphorylation levels of a-ERM-Thr567were increased more than2-fold compared to those in DMSO-treated cells,which con?rmed that emodin rescued the phosphorylation of ezrin(Fig.4).Emodin treatment in DLD-1 cells did not affect the phosphorylation of a-ERM-Thr567.

In summary,during a search for PRL-3modulators from natural sources,we identi?ed emodin as a PRL-3phosphatase inhibitor. Emodin inhibited PRL-3-induced cell migration and invasion in DLD-1tumor cells and rescued the phosphorylation of ezrin.Col-lectively,our results indicate the involvement of PRL-3in promot-ing tumor cell migration and invasion.Notably,we showed that the inhibition of migration and invasion of DLD-1(PRL-3)cells were associated with inhibition of PRL-3phosphatase activity by emodin.Therefore,emodin is a lead molecule for discovery of the PRL-3phosphatase inhibitor.

Acknowledgements

This work was supported by Woosuk University(2011)(The Re-gional Core Research Program/Center for Healthcare Technology Development)and the KRIBB Research Initiative Program.This work was also supported by the National Research Foundation of Korea(NRF-2011-0015768)and the National Chemical Genomics Research Program grant funded from the Korea government (MEST).

References and notes

1.Steeg,P.S.Nature Med.2006,12,895.

2.Nguyen,D.X.;Massague,J.Nature Rev.Genet.2007,8,341.

3.Christofori,G.Nature2006,441,44

4.

4.Chiang,A.C.;Massague,J.New Engl.J.Med.2008,359,2814–2823.

5.Fidler,I.J.Nature Rev.Cancer2003,3,453.

6.Klein,C.A.Nature Rev.Cancer2009,9,302.

7.Jemal,A.;Bray,F.;Center,M.M.;Ferlay,J.;Ward,E.;Forman,D.C.A.Cancer J.

Clin.2011,61(69),2011.

8.Jemal,A.;Siegel,R.;Xu,J.;Ward,E.Cancer J.Clin.2010,60,277.

9.Anaya,D.A.;Becker,N.S.;Abraham,N.S.Crit.Rev.Oncol.Hematol.2011,77,

100.

10.Zeng,Q.;Si,X.;Horstmann,H.;Xu,Y.;Hong,W.;Pallen,C.J.J.Biol.Chem.2000,

275,21444.

11.Saha,S.;Bardelli,A.;Buckhaults,P.;Velculescu,V.E.;Rago,C.;Croix,B.S.;

Romans,K.E.;Choti,M.A.;Lengauer,C.;Kinzler,K.W.;Vogelstein,B.Science 2001,294,1343.

12.Bardelli,A.;Saha,S.;Sager,J.A.;Romans,K.E.;Xin,B.;Markowitz,S.D.;

Lengauer,C.;Velculescu,V.E.;Kinzler,K.W.;Vogelstein,B.Clin.Cancer Res.

2003,9,5607.

13.Ooki,A.;Yamashita,K.;Kikuchi,S.;Sakuramoto,S.;Katada,N.;Watanabe,M.

Int.J.Cancer2010,127,543.

14.Zeng,Q.;Dong,J.M.;Guo,K.;Li,J.;Tan,H.X.;Koh,V.;Pallen,C.J.;Manser,E.;

Hong,W.Cancer Res.2003,63,2716.

15.Guo,K.;Li,J.;Wang,H.;Osato,M.;Tang,J.P.;Quah,S.Y.;Gan,B.Q.;Zeng,Q.

Cancer Res.2006,66,9625.

16.Guo,K.;Li,J.;Tang,J.P.;Koh,V.;Gan,B.Q.;Zeng,Q.Cancer Biol.Ther.2004,3,

945.

17.Choi,S.K.;Oh,H.M.;Lee,S.K.;Jeong,D.G.;Ryu,S.E.;Son,K.H.;Han,D.C.;

Sung,N.D.;Baek,N.I.;Kwon,B.M.Nat.Prod.Res.2006,20,341.

18.Moon,M.K.;Han,Y.M.;Lee,Y.J.;Lee,L.H.;Yang,J.H.;Kwon,B.M.;Kim,D.K.

Arch.Pharm.Res.2010,33,1747.

19.Huang,Q.;Lu,G.;Shen,H.M.;Chung,M.C.M.;Ong,C.N.Med.Res.Rev.2007,

27,609.

20.PRL-3assay was performed using a DiFMUP assay(6,8-di?uoro-4-

methylumbelliferyl phosphate).The assays were performed in96-well plates using a reaction mixture containing20mM Tris–HCl,pH8.0,10mM DTT,

0.01%Triton X-100and4mM DiFMUP.A total volume of200l L was used in

the assay.Then10l L of PRL-3puri?ed protein was added to190l L of reaction mixture.The mixture was incubated for1h at room temperature and the reaction was stopped by adding of2mM sodium orthovanadate.The results were determined using a?uorimeter,at an excitation/emission wavelength of 355nm/460nm.

21.Daouti,S.;Li,W.H.;Qian,H.;Huang,K.S.;Holmgren,J.;Levin,W.;Reik,L.;

McGady,D.L.;Gillespie,P.;Perrotta,A.;Bian,H.;Reidhaar-Olson,J.F.;Bliss,S.

A.;Olivier, A.R.;Sergi,J. A.;Fry, D.;Danho,W.;Ritland,S.;Fotouhi,N.;

Heimbrook,D.;Niu,H.Cancer Res.2008,68,1162.

22.Rouleau,C.;Roy,A.;Martin,T.S.;Dufault,M.R.;Boutin,P.;Liu,D.;Zhang,M.;

Puorro-Radzwill,K.;Rulli,L.;Reczek,D.;Bagley,R.;Byrne,A.;Weber,W.;

Roberts, B.;Klinger,K.;Brondyk,W.;Nacht,M.;Madden,S.;Burrier,R.;

Shankara,S.;Teicher,B.A.Mol.Cancer Ther.2006,5,219.

23.Generation of stable DLD-1cell lines expressing PRL-3and mutant PRL-3.The

cDNA containing full-length PRL-3gene was obtained from human brain Quick-Clone cDNA(Clontech).The forward primer(50-CGAATTCATGGCTCGGATGAAC-

30)and the reverse primer(50-CCGCTCGAGCGGCCTGAGCTACATAAC-30)were

used for ampli?cation of PRL-3gene.The PCR fragments were digested with EcoR1and Xho,and were inserted into the modi?ed pcDNA3plasmid which has the HA epitope(10amino acid residues)at the NH2termini.The expression plasmid was con?rmed using DNA sequencing analysis and introduced into DLD-1human colon adenocarcinoma cells(American Type Culture Collection, Manassas,VA)using Lipofectamine2000(Invitrogen,USA).The DLD-1cell was cultured in RPMI1640supplemented with10%FBS.Stable cells were selected using culture media containing1mg/ml geneticin(G418,Invitrogen,USA)for 20–30days.Stable transformants were continuously maintained in RPMI1640 media containing200l g/ml Geneticin.

24.Migration and Invasion Assays in Boyden Chambers.A cell migration assay was

performed using8.0l m pore size Transwell inserts(Falcon,USA).Cells were harvested with trypsin–EDTA,and washed twice with serum-free RPMI1640 medium.Cells were resuspended in serum-free medium and8?104cells in

0.2ml Serum free RPMI1640medium were added to the upper chamber.

Various concentrations of chemicals in0.5ml RPMI1640medium with10% FBS were placed into the lower chamber.Cells were incubated for15or20h at 37°C in a humidi?ed atmosphere of5%CO2.At the end of the experiment,the cells on the upper surface of the membrane were removed using cotton swabs.

The migrant cells that were attached to the lower surface were stained with crystal violet and incubated for30min.The membrane was washed several times with phosphate buffered saline,and the cells has penetrated the?lter were counted under a microscope.

For the invasion assay,?lters were precoated on the upper side with Matrigel (20l g per well;BD Biosciences).Serum-free conditioned medium from DLD-1 (PRL-3)cells cultured for20h was placed on the lower compartment of the chamber.DLD-1-Empty vector and DLD-1(PRL-3)cells were prestained with calcein AM(5ug/ml;Molecular Probes)for30min at37°C,and105cells were seeded in serum-free MEM on the upper compartment of the chambers for6h at37°C.The?uorescence of the cells on the lower side of the?lter was measured in the wells using a computer-based?uorescence reader(Fusion;

Packard).All conditions were performed in triplicate.Values shown are the mean±SEM.

25.Forte,E.;Orsatti,L.;Talamo,F.;Barbato,G.;De Francesco,R.;Tomei,L.Biochim.

Biophys.Acta2008,1783,334.

326Y.-M.Han et al./Bioorg.Med.Chem.Lett.22(2012)323–326